All charts
www.databrew.cc
2020-03-23
## Load libraries
library(covid19)
library(ggplot2)
library(lubridate)
library(dplyr)
library(ggplot2)
library(sp)
library(raster)
library(viridis)
library(ggthemes)
library(sf)
library(rnaturalearth)
library(rnaturalearthdata)
library(RColorBrewer)Italy regions, Spanish regions, Chinese regions (adjusted for population)
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
# Italian data
b <- ita %>%
left_join(ita_pop) %>%
mutate(country = 'Italy')
# Chinese data
d <- df %>% filter(country == 'China') %>%
mutate(cases = confirmed_cases) %>%
mutate(ccaa = district) %>%
mutate(country = 'China') %>%
left_join(chi_pop)
# join
joined <- bind_rows(a, b, d)
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
# Define plot data
pd <- joined %>% filter(days_since_start_deaths_pm >= 0) %>%
mutate(country = ifelse(country == 'China',
'Hubei (China)',
ifelse(country == 'Italy', 'Italia', 'España')))
# Define label data
label_data <- pd %>% group_by(ccaa) %>% filter(
(
(country == 'Hubei (China)' & days_since_start_deaths_pm == 22) |
(date == max(date) & country == 'España' & deaths_pm > 25) |
(date == max(date) & country == 'Italia' &
ccaa != 'Liguria' & days_since_start_deaths_pm > 10)
))
# Get differential label data based on what to be emphasized
label_data_big <- label_data %>%
filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
label_data_small <- label_data %>%
filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
pd_big <- pd %>%
filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
pd_small <- pd %>%
filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
# cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Set2'))(length(unique(pd$country)))
# cols <- rainbow(3)
cols <- c('darkred', '#FF6633', '#006666')
ggplot(data = pd_big,
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.9,
size = 2) +
geom_line(data = pd_small,
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
color = country),
alpha = 0.7,
size = 1) +
scale_y_log10() +
scale_color_manual(name = '',
values = c(cols)) +
theme_simple() +
theme(legend.position = 'top') +
labs(x = 'Dias desde "el comienzo del brote"',
y = 'Muertes por millón de habitantes',
title = 'Muertes por 1.000.000 habitantes',
subtitle = paste0('Dia 0 ("comienzo del brote") = primer día a ', x_deaths_pm, ' o más muertes acumuladas por milión de población\nLíneas rojas: CCAA; líneas verde-azules: regiones italianas; línea naranja: Hubei, China'),
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data_big,
aes(x = days_since_start_deaths_pm + 0.6,
y = (deaths_pm + 20),
label = gsub(' ', '\n', ccaa),
color = country),
size = 8,
alpha = 0.9,
show.legend = FALSE) +
geom_text(data = label_data_small,
aes(x = days_since_start_deaths_pm + 0.6,
y = deaths_pm ,
label = gsub(' ', '\n', ccaa),
color = country),
size = 5,
alpha = 0.6,
show.legend = FALSE) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20),
legend.text = element_text(size = 16),
plot.title = element_text(size = 30)) +
xlim(0, 26)
ggsave('~/Desktop/china_spain_italy_comparison.png',
height = 7,
width = 10)Italy regions, Spanish regions, Chinese regions (raw numbers, not adjusted for population)
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
# Italian data
b <- ita %>%
left_join(ita_pop) %>%
mutate(country = 'Italy')
# Chinese data
d <- df %>% filter(country == 'China') %>%
mutate(cases = confirmed_cases) %>%
mutate(ccaa = district) %>%
mutate(country = 'China') %>%
left_join(chi_pop)
# join
joined <- bind_rows(a, b, d)
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
# Define plot data
pd <- joined %>% filter(days_since_start_deaths >= 0) %>%
mutate(country = ifelse(country == 'China',
'China',
ifelse(country == 'Italy', 'Italia', 'España')))
# Define label data
label_data <- pd %>% group_by(ccaa) %>% filter(
(
(country == 'China' & deaths >10 & days_since_start_deaths == 29) |
(date == max(date) & country == 'España' & deaths > 90) |
(date == max(date) & country == 'Italia' &
ccaa != 'Liguria' & days_since_start_deaths > 10)
))
# Get differential label data based on what to be emphasized
label_data_big <- label_data %>%
filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
label_data_small <- label_data %>%
filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
pd_big <- pd %>%
filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
pd_small <- pd %>%
filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
# cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Set2'))(length(unique(pd$country)))
# cols <- rainbow(3)
cols <- c( '#FF6633', 'darkred', '#006666')
ggplot(data = pd_big,
aes(x = days_since_start_deaths,
y = deaths,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.9,
size = 2) +
geom_line(data = pd_small,
aes(x = days_since_start_deaths,
y = deaths,
color = country),
alpha = 0.7,
size = 1) +
scale_y_log10() +
scale_color_manual(name = '',
values = c(cols)) +
theme_simple() +
theme(legend.position = 'top') +
labs(x = 'Dias desde el primer día con 5 o más muertes acumuladas',
y = 'Muertes',
title = 'Muertes por COVID-19',
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data_big,
aes(x = days_since_start_deaths + 1.6,
y = ifelse(ccaa == 'Hubei', (deaths -500),
ifelse(ccaa == 'Lombardia', (deaths + 700),
(deaths + 300))),
label = gsub(' ', '\n', ccaa),
color = country),
size = 8,
alpha = 0.9,
show.legend = FALSE) +
geom_text(data = label_data_small,
aes(x = days_since_start_deaths + 1.6,
align = 'left',
y = deaths ,
label = ccaa,
# label = gsub(' ', '\n', ccaa),
color = country),
size = 5,
alpha = 0.6,
show.legend = FALSE) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20),
legend.text = element_text(size = 16),
plot.title = element_text(size = 30)) +
xlim(0, 35)
ggsave('~/Desktop/china_spain_italy_comparison_raw.png',
height = 7,
width = 10) ANIMATION: Italy regions, Spanish regions, Chinese regions (raw numbers, not adjusted for population)
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
# Italian data
b <- ita %>%
left_join(ita_pop) %>%
mutate(country = 'Italy')
# Chinese data
d <- df %>% filter(country == 'China') %>%
mutate(cases = confirmed_cases) %>%
mutate(ccaa = district) %>%
mutate(country = 'China') %>%
left_join(chi_pop)
# join
joined <- bind_rows(a, b, d)
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
# Define plot data
pd <- joined %>% filter(days_since_start_deaths >= 0) %>%
mutate(country = ifelse(country == 'China',
'China',
ifelse(country == 'Italy', 'Italia', 'España')))
add_zero <- function(x, n){
x <- as.character(x)
adders <- n - nchar(x)
adders <- ifelse(adders < 0, 0, adders)
for (i in 1:length(x)){
if(!is.na(x[i])){
x[i] <- paste0(
paste0(rep('0', adders[i]), collapse = ''),
x[i],
collapse = '')
}
}
return(x)
}
# # Define label data
# label_data <- pd %>% group_by(ccaa) %>% filter(
# (
# (country == 'China' & deaths >10 & days_since_start_deaths == 29) |
# (date == max(date) & country == 'España' & deaths > 90) |
# (date == max(date) & country == 'Italia' &
# ccaa != 'Liguria' & days_since_start_deaths > 10)
# ))
# # Get differential label data based on what to be emphasized
# label_data_big <- label_data %>%
# filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
# label_data_small <- label_data %>%
# filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
#
pd_big <- pd %>%
filter(ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
pd_small <- pd %>%
filter(!ccaa %in% c('Madrid', 'Lombardia', 'Hubei'))
# cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Set2'))(length(unique(pd$country)))
# cols <- rainbow(3)
cols <- c( '#FF6633', 'darkred', '#006666')
the_dir <- '~/Desktop/animation/'
dir.create(the_dir)
the_dates <- sort(unique(c(pd_big$date, pd_small$date)))
for(i in 1:length(the_dates)){
the_date <- the_dates[i]
pd_big_sub <- pd_big %>% filter(date <= the_date)
pd_big_current <- pd_big_sub %>% filter(date == the_date)
pd_small_sub <- pd_small %>% filter(date <= the_date)
pd_small_current <- pd_small_sub %>% filter(date == the_date)
label_data_big <-
pd_big_sub %>%
filter(ccaa %in% c('Lombardia', 'Madrid', 'Hubei')) %>%
group_by(ccaa) %>%
filter(date == max(date)) %>%
ungroup %>%
mutate(days_since_start_deaths = ifelse(ccaa == 'Hubei' &
days_since_start_deaths >32,
32,
days_since_start_deaths))
label_data_small <-
pd_small_sub %>%
filter(ccaa %in% c('Emilia Romagna',
'Cataluña',
'CLM',
'País Vasco',
'Veneto',
'Piemonte',
'Henan',
'Heilongjiang')) %>%
group_by(ccaa) %>%
filter(date == max(date))
n_countries <- length(unique(pd_big_sub$country))
if(n_countries == 3){
sub_cols <- cols
}
if(n_countries == 2){
sub_cols <- cols[c(1,3)]
}
if(n_countries == 1){
sub_cols <- cols[1]
}
g <- ggplot(data = pd_big_sub,
aes(x = days_since_start_deaths,
y = deaths,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.9,
size = 2) +
geom_line(data = pd_small_sub,
aes(x = days_since_start_deaths,
y = deaths,
color = country),
alpha = 0.7,
size = 1) +
geom_point(data = pd_big_current,
aes(x = days_since_start_deaths,
y = deaths,
color = country),
size = 3) +
geom_point(data = pd_small_current,
aes(x = days_since_start_deaths,
y = deaths,
color = country),
size = 1, alpha = 0.6) +
scale_y_log10(limits = c(5, 3250)) +
scale_color_manual(name = '',
values = sub_cols) +
theme_simple() +
theme(legend.position = 'top') +
labs(x = 'Dias desde el primer día con 5 o más muertes acumuladas',
y = 'Muertes',
title = format(the_date, '%d %b'),
subtitle = 'Muertes por COVID-19',
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data_big,
aes(x = days_since_start_deaths + 1,
y = deaths,
hjust = 0,
label = gsub(' ', '\n', ccaa),
color = country),
size = 8,
alpha = 0.9,
show.legend = FALSE) +
geom_text(data = label_data_small,
aes(x = days_since_start_deaths + 1.6,
y = deaths ,
label = ccaa,
# label = gsub(' ', '\n', ccaa),
color = country),
size = 5,
alpha = 0.6,
show.legend = FALSE) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20),
legend.text = element_text(size = 16),
plot.title = element_text(size = 35),
plot.subtitle = element_text(size = 24)) +
xlim(0, 36)
message(i)
ggsave(paste0(the_dir, add_zero(i, 3), '.png'),
height = 7,
width = 10.5)
}# Command line
cd ~/Desktop/animation
mogrify -resize 50% *.png
convert -delay 10 -loop 0 *.png result.gif
ANIMATION: Spain only
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
joined <- a
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
# Define plot data
pd <- joined %>% filter(days_since_start_deaths >= 0) %>%
mutate(country = ifelse(country == 'China',
'China',
ifelse(country == 'Italy', 'Italia', 'España')))
add_zero <- function(x, n){
x <- as.character(x)
adders <- n - nchar(x)
adders <- ifelse(adders < 0, 0, adders)
for (i in 1:length(x)){
if(!is.na(x[i])){
x[i] <- paste0(
paste0(rep('0', adders[i]), collapse = ''),
x[i],
collapse = '')
}
}
return(x)
}
bigs <- c('Madrid', 'Cataluña', 'CLM', 'CyL', 'País Vasco', 'La Rioja')
pd_big <- pd %>%
filter(ccaa %in% bigs)
pd_small <- pd %>%
filter(!ccaa %in% bigs)
# cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Set2'))(length(unique(pd$country)))
# cols <- rainbow(3)
cols <- colorRampPalette(c('#A16928','#bd925a','#d6bd8d','#edeac2', '#b5c8b8','#79a7ac','#2887a1'))(length(unique(pd$country)))
the_dir <- '~/Desktop/animation/'
dir.create(the_dir)
the_dates <- sort(unique(c(pd_big$date, pd_small$date)))
for(i in 1:length(the_dates)){
the_date <- the_dates[i]
pd_big_sub <- pd_big %>% filter(date <= the_date)
pd_big_current <- pd_big_sub %>% filter(date == the_date)
pd_small_sub <- pd_small %>% filter(date <= the_date)
pd_small_current <- pd_small_sub %>% filter(date == the_date)
label_data_big <-
pd_big_sub %>%
filter(ccaa %in% bigs) %>%
group_by(ccaa) %>%
filter(date == max(date)) %>%
ungroup
label_data_small <-
pd_small_sub %>%
group_by(ccaa) %>%
filter(date == max(date))
# sub_cols <- colorRampPalette(c('#A16928','#bd925a','#d6bd8d','#edeac2', '#b5c8b8','#79a7ac','#2887a1'))(length(unique(pd$ccaa)))
sub_cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 8, name = 'Dark2'))(length(unique(pd$ccaa)))
# sub_cols <- rainbow((length(unique(pd$ccaa))))
g <- ggplot(data = pd_big_sub,
aes(x = days_since_start_deaths,
y = deaths,
group = ccaa)) +
geom_line(aes(color = ccaa),
alpha = 0.9,
size = 2) +
geom_line(data = pd_small_sub,
aes(x = days_since_start_deaths,
y = deaths,
color = ccaa),
alpha = 0.7,
size = 1) +
geom_point(data = pd_big_current,
aes(x = days_since_start_deaths,
y = deaths,
color = ccaa),
size = 3) +
geom_point(data = pd_small_current,
aes(x = days_since_start_deaths,
y = deaths,
color = ccaa),
size = 1, alpha = 0.6) +
geom_point(data = pd,
aes(x = days_since_start_deaths,
y = deaths,
color = ccaa),
size = 1, alpha = 0.01) +
scale_y_log10(limits = c(1, 1100),
breaks = c(10, 50, 100, 500, 1000)) +
scale_color_manual(name = '',
values = sub_cols) +
theme_simple() +
theme(legend.position = 'top') +
labs(x = 'Dias desde el primer día con 5 o más muertes acumuladas',
y = 'Muertes',
title = format(the_date, '%d %b'),
subtitle = 'Muertes por COVID-19',
caption = '@joethebrew | www.databrew.cc') +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20),
legend.text = element_text(size = 16),
plot.title = element_text(size = 35),
plot.subtitle = element_text(size = 24)) +
xlim(0, 20) +
theme(legend.position = 'none')
message(i)
if(nrow(label_data_big) > 0){
g <- g +
geom_text(data = label_data_big,
aes(x = days_since_start_deaths + 0.2,
y = deaths,
hjust = 0,
label = gsub(' ', ' ', ccaa),
color = ccaa),
size = 8,
alpha = 0.9,
show.legend = FALSE) +
geom_text(data = label_data_small,
aes(x = days_since_start_deaths + 0.2,
y = deaths ,
label = ccaa,
# label = gsub(' ', '\n', ccaa),
color = ccaa),
size = 5,
alpha = 0.6,
show.legend = FALSE)
}
ggsave(paste0(the_dir, add_zero(i, 3), '.png'),
height = 7,
width = 12)
}# Command line
cd ~/Desktop/animation
mogrify -resize 50% *.png
convert -delay 25 -loop 0 *.png result.gif
Italy regions for Spanish regions
# Spanish data
a <- esp_df %>%
left_join(esp_pop) %>%
mutate(country = 'Spain')
# Italian data
b <- ita %>%
left_join(ita_pop) %>%
mutate(country = 'Italy')
# join
joined <- bind_rows(a, b)
# Get deaths per milllion
joined$deaths_pm <- joined$deaths / joined$pop * 1000000
joined$cases_pm <- joined$cases / joined$pop * 1000000
# Get the days since paradigm
x_deaths <- 5
x_deaths_pm <- 5
x_cases <- 50
x_cases_pm <- 50
joined <- joined %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(start_deaths = min(date[deaths >= x_deaths]),
start_cases = min(date[cases >= x_cases]),
start_deaths_pm = min(date[deaths_pm >= x_deaths_pm]),
start_cases_pm = min(date[cases_pm >= x_cases_pm])) %>%
ungroup %>%
mutate(days_since_start_deaths = date - start_deaths,
days_since_start_cases = date - start_cases,
days_since_start_deaths_pm = date - start_deaths_pm,
days_since_start_cases_pm = date - start_cases_pm)
ggplot(data = joined %>% filter(days_since_start_deaths_pm >= 0),
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.8,
size = 2) +
scale_y_log10() +
scale_color_manual(name = '',
values = c('darkorange', 'purple')) +
theme_simple() +
theme(legend.position = 'none') +
labs(x = 'Days since "start out outbreak"',
y = 'Deaths per million',
title = 'Deaths per capita, Italian regions vs. Spanish autonomous communities',
subtitle = paste0('Day 0 ("start of outbreak") = first day at ', x_deaths_pm, ' or greater cumulative deaths per million'),
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = joined %>% group_by(ccaa) %>% filter(date == max(date) &
(
(country == 'Spain' & deaths_pm > 25) |
(country == 'Italy' & days_since_start_deaths_pm > 10)
)),
aes(x = days_since_start_deaths_pm + 0.6,
y = deaths_pm,
label = gsub(' ', '\n', ccaa),
color = country),
size = 6) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20)) +
xlim(0, 21)
ggsave('~/Desktop/italy_comparison.png',
height = 6,
width = 10)
# Separate for Catalonia
pd <- joined %>% filter(days_since_start_deaths_pm >= 0) %>%
mutate(country = ifelse(ccaa == 'Cataluña',
'Catalonia',
country)) %>%
mutate(ccaa = ifelse(ccaa == 'Cataluña', 'Catalunya', ccaa))
pdcat <- pd %>% filter(country == 'Catalonia')
label_data <- pd %>% group_by(ccaa) %>% filter(date == max(date) &
(
(country == 'Catalonia') |
(country == 'Spain' & deaths_pm > 25) |
(country == 'Italy' & days_since_start_deaths_pm > 10)
))
ggplot(data = pd,
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.3,
size = 1.5) +
geom_line(data = pdcat,
aes(color = country),
alpha = 0.8,
size = 2) +
scale_y_log10() +
scale_color_manual(name = '',
values = c('darkred', 'darkorange', "purple")) +
theme_simple() +
theme(legend.position = 'none') +
labs(x = 'Dies des del "començament del brot"',
y = 'Morts per milió',
title = 'Morts per càpita: Catalunya, comunitats autònomes, regions italianes',
subtitle = paste0('Dia 0 ("començament del brot") = primer dia a ', x_deaths_pm, ' o més morts acumulades per milió de població'),
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data,
aes(x = days_since_start_deaths_pm + 0.6,
y = deaths_pm +3,
label = gsub(' ', '\n', ccaa),
color = country),
size = 6,
alpha = 0.7) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20)) +
xlim(0, 21)
ggsave('~/Desktop/cat_italy_comparison.png',
height = 6,
width = 10)
# Straightforward Lombardy, Madrid, Cat comparison
specials <- c('Lombardia', 'Madrid')
pd <- joined %>% filter(days_since_start_deaths_pm >= 0) %>%
mutate(country = ifelse(ccaa == 'Cataluña',
'Catalonia',
country)) %>%
mutate(ccaa = ifelse(ccaa == 'Cataluña', 'Catalunya', ccaa))
pdcat <- pd %>% filter(ccaa %in% specials)
label_data <- pd %>% group_by(ccaa) %>% filter(date == max(date) &
(
# (country == 'Catalonia') |
(country == 'Spain' & deaths_pm > 20) |
(country == 'Italy' & days_since_start_deaths_pm >= 10)
))
ggplot(data = pd,
aes(x = days_since_start_deaths_pm,
y = deaths_pm,
group = ccaa)) +
geom_line(aes(color = country),
alpha = 0.3,
size = 1.5) +
geom_line(data = pdcat,
aes(color = country),
alpha = 0.8,
size = 2) +
scale_y_log10() +
scale_color_manual(name = '',
values = c('darkred', 'darkorange', "purple")) +
theme_simple() +
theme(legend.position = 'none') +
labs(x = 'Dias desde "el comienzo del brote"',
y = 'Muertes por millón de habitantes',
title = 'Muertes acumuladas por 1.000.000 habitantes',
subtitle = paste0('Dia 0 ("comienzo del brote") = primer día a ', x_deaths_pm, ' o más muertes acumuladas por milión de población'),
caption = '@joethebrew | www.databrew.cc') +
geom_text(data = label_data %>% filter(!ccaa %in% specials),
aes(x = days_since_start_deaths_pm + 0.4,
y = deaths_pm +3,
label = gsub(' ', '\n', ccaa),
color = country),
size = 5,
alpha = 0.5) +
geom_text(data = label_data %>% filter(ccaa %in% specials),
aes(x = days_since_start_deaths_pm ,
y = deaths_pm +30,
label = gsub(' ', '\n', ccaa),
color = country),
size = 8,
alpha = 0.8) +
theme(axis.text = element_text(size = 14),
axis.title = element_text(size = 20)) +
xlim(0, 19)
ggsave('~/Desktop/mad_lom_italy_comparison.png',
height = 6,
width = 10)Africa
african_isos <- world_pop %>% filter(region == 'Africa')
pd <- df_country %>%
group_by(iso) %>%
filter(max(confirmed_cases) > 0) %>%
filter(iso %in% african_isos$iso) %>%
mutate(country = ifelse(country == 'Central African Republic', 'C.A.R.',
ifelse(grepl('Brazza', country), 'Congo, Rep.',
ifelse(grepl('Kinsha', country), 'Congo, D.R.',
ifelse(grepl('Equatorial', country),
'Equat. Guinea', country)))))
cols <- colorRampPalette(brewer.pal(n = 8, 'Spectral'))(length(unique(pd$country)))
cols <- sample(cols, length(cols))
# Plot
ggplot(data = pd,
aes(x = date,
# color = country,
# fill = country,
y = confirmed_cases)) +
theme_simple() +
# geom_point() +
# geom_line() +
geom_area(fill = 'darkred', alpha = 0.3, color = 'darkred') +
# scale_color_manual(name = '',
# values = cols) +
# scale_fill_manual(name = '',
# values = cols) +
theme(legend.position = 'none',
axis.text = element_text(size = 5),
legend.text = element_text(size = 6)) +
# scale_y_log10() +
facet_wrap(~country,
scales = 'free') +
labs(x = '',
y = 'Confirmed cases',
title = 'COVID-19 in Africa')
ggsave('~/Desktop/africa.pdf',
height = 8,
width = 12)Latin America
latam_isos <- world_pop %>% filter(sub_region == 'Latin America and the Caribbean')
pd <- df_country %>%
group_by(iso) %>%
filter(max(confirmed_cases) > 0) %>%
filter(iso %in% latam_isos$iso)
cols <- colorRampPalette(brewer.pal(n = 8, 'Spectral'))(length(unique(pd$country)))
cols <- sample(cols, length(cols))
# Plot
ggplot(data = pd,
aes(x = date,
# color = country,
# fill = country,
y = confirmed_cases)) +
theme_simple() +
# geom_point() +
# geom_line() +
geom_area(fill = 'darkred', alpha = 0.3, color = 'darkred') +
# scale_color_manual(name = '',
# values = cols) +
# scale_fill_manual(name = '',
# values = cols) +
theme(legend.position = 'none',
axis.text = element_text(size = 5),
legend.text = element_text(size = 6)) +
# scale_y_log10() +
facet_wrap(~country,
scales = 'free') +
labs(x = '',
y = 'Confirmed cases',
title = 'COVID-19 in Latin America and the Caribbean')
ggsave('~/Desktop/latam.pdf',
height = 8,
width = 12)Rolling average new events
roll_curve <- function(data,
n = 4,
deaths = FALSE,
scales = 'fixed',
nrow = NULL,
ncol = NULL,
pop = FALSE){
# Create the n day rolling avg
ma <- function(x, n = 5){
if(length(x) >= n){
stats::filter(x, rep(1 / n, n), sides = 1)
} else {
new_n <- length(x)
stats::filter(x, rep(1 / new_n, new_n), sides = 1)
}
}
pd <- data
if(deaths){
pd$var <- pd$deaths_non_cum
} else {
pd$var <- pd$confirmed_cases_non_cum
}
if('ccaa' %in% names(pd)){
pd$geo <- pd$ccaa
} else {
pd$geo <- pd$iso
}
if(pop){
# try to get population
if(any(pd$geo %in% unique(esp_df$ccaa))){
right <- esp_pop
right_var <- 'ccaa'
} else {
right <- world_pop
right_var <- 'iso'
}
pd <- pd %>% left_join(right %>% dplyr::select(all_of(right_var), pop),
by = c('geo' = right_var)) %>%
mutate(var = var / pop * 100000)
}
pd <- pd %>%
arrange(date) %>%
group_by(geo) %>%
mutate(with_lag = ma(var, n = n))
ggplot() +
geom_bar(data = pd,
aes(x = date,
y = var),
stat = 'identity',
fill = 'grey',
alpha = 0.8) +
geom_area(data = pd,
aes(x = date,
y = with_lag),
color = 'red',
fill = 'red',
alpha = 0.6) +
facet_wrap(~geo, scales = scales, nrow = nrow, ncol = ncol) +
theme_simple() +
labs(x = '',
y = ifelse(deaths, 'Deaths', 'Cases'),
title = paste0('Daily (non-cumulative) ', ifelse(deaths, 'deaths', 'cases'),
ifelse(pop, ' (per 100,000)', '')),
subtitle = paste0('Data as of ', max(data$date),
'.\nRed line = ', n, ' day rolling average. Grey bar = day-specific value.')) +
theme(axis.text.x = element_text(size = 7,
angle = 90,
hjust = 0.5, vjust = 1)) #+
# scale_x_date(name ='',
# breaks = sort(unique(pd$date)),
# labels = format(sort(unique(pd$date)), '%b %d'))
# scale_y_log10()
}plot_data <- df_country %>% filter(country %in% c('Spain', 'France', 'Italy', 'Germany', 'Belgium', 'Norway')) %>% mutate(geo = country)
roll_curve(plot_data, pop = T)
dir.create('~/Desktop/countries')
roll_curve_country <- function(the_country = 'Spain'){
plot_data <- df_country %>% filter(country %in% the_country) %>% mutate(geo = country)
g1 <- roll_curve(plot_data, pop = F)
g2 <- roll_curve(plot_data, pop = T)
g3 <- roll_curve(plot_data, pop = F, deaths = T)
g4 <- roll_curve(plot_data, pop = T, deaths = T)
ggsave(paste0('~/Desktop/countries/', the_country, '1.png'), g1)
ggsave(paste0('~/Desktop/countries/', the_country, '2.png'), g2)
ggsave(paste0('~/Desktop/countries/', the_country, '3.png'), g3)
ggsave(paste0('~/Desktop/countries/', the_country, '4.png'), g4)
}
countries <- c('Spain', 'France', 'Italy', 'Germany', 'Belgium', 'Norway', 'US', 'United Kingdom', 'Korea, South',
'China', 'Japan', 'Switzerland', 'Sweden', 'Denmark', 'Netherlands', 'Iran', 'Canada')
for(i in 1:length(countries)){
roll_curve_country(the_country = countries[i])
}plot_data <- df_country %>% filter(country == 'Spain') %>% mutate(geo = country)
roll_curve(plot_data)
plot_data <- esp_df %>% mutate(geo = ccaa)
roll_curve(plot_data, pop = T, deaths = T)
plot_data <- df_country %>% filter(country == 'Spain') %>% mutate(geo = country)
roll_curve(plot_data, deaths = T)
# Latest in Spain
pd <- esp_df %>%
filter(date == max(date)) %>%
mutate(p = deaths / sum(deaths) * 100)
text_size <- 12
# deaths
ggplot(data = pd,
aes(x = ccaa,
y = deaths)) +
geom_bar(stat = 'identity',
fill = 'black') +
theme_simple() +
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
labs(x = '',
y = 'Deaths | Muertes',
title = 'COVID-19 deaths in Spain',
subtitle = paste0('Data as of ', max(pd$date)),
caption = 'github.com/databrew/covid19 | joe@databrew.cc') +
theme(legend.position = 'top',
legend.text = element_text(size = text_size * 2),
axis.title = element_text(size = text_size * 2),
plot.title = element_text(size = text_size * 2.3),
axis.text.x = element_text(size = text_size * 1.5)) +
geom_text(data = pd %>% filter(deaths > 0),
aes(x = ccaa,
y = deaths,
label = paste0(deaths, '\n(',
round(p, digits = 1), '%)')),
size = text_size * 0.3,
nudge_y = 80) +
ylim(0, 1200)
ggsave('~/Desktop/spain.png')Muertes relativas por CCAA
# Latest in Spain
pd <- esp_df %>%
filter(date == max(date)) %>%
mutate(p = deaths / sum(deaths) * 100)
pd <- pd %>% left_join(esp_pop)
text_size <- 12
pd$value <- pd$deaths / pd$pop * 100000
# deaths
ggplot(data = pd,
aes(x = ccaa,
y = value)) +
geom_bar(stat = 'identity',
fill = 'black') +
theme_simple() +
theme(axis.text.x = element_text(angle = 90, hjust = 1, vjust = 0.5)) +
labs(x = '',
y = 'Deaths per 100,000',
title = 'COVID-19 deaths per 100.000',
subtitle = paste0('Data as of ', max(pd$date)),
caption = 'github.com/databrew/covid19 | joe@databrew.cc') +
theme(legend.position = 'top',
legend.text = element_text(size = text_size * 2),
axis.title = element_text(size = text_size * 2),
plot.title = element_text(size = text_size * 2.3),
axis.text.x = element_text(size = text_size * 1.5)) +
geom_text(data = pd %>% filter(value > 0),
aes(x = ccaa,
y = value,
label = paste0(round(value, digits = 2), '\n(',
deaths, '\ndeaths)')),
size = text_size * 0.3,
nudge_y = 2.5) +
ylim(0, 19)
ggsave('~/Desktop/spai2.png')plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(!ccaa %in% c('Melilla'))
roll_curve(plot_data, scales = 'fixed')
ggsave('~/Desktop/a.png',
width = 1280 / 150,
height = 720 / 150)Loop for everywhere (see desktop)
dir.create('~/Desktop/ccaas')
ccaas <- sort(unique(esp_df$ccaa))
for(i in 1:length(ccaas)){
this_ccaa <- ccaas[i]
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(ccaa == this_ccaa)
roll_curve(plot_data, scales = 'fixed') + theme(strip.text = element_text(size = 30))
ggsave(paste0('~/Desktop/ccaas/', i, this_ccaa, '_cases.png'),
width = 1280 / 150,
height = 720 / 150)
}
ccaas <- sort(unique(esp_df$ccaa))
for(i in 1:length(ccaas)){
this_ccaa <- ccaas[i]
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(ccaa == this_ccaa)
roll_curve(plot_data, scales = 'fixed', pop = TRUE) + theme(strip.text = element_text(size = 30))
ggsave(paste0('~/Desktop/ccaas/', i, this_ccaa, '_cases_pop.png'),
width = 1280 / 150,
height = 720 / 150)
}
# Deaths too
for(i in 1:length(ccaas)){
this_ccaa <- ccaas[i]
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(ccaa == this_ccaa)
roll_curve(plot_data, deaths = T, scales = 'fixed') + theme(strip.text = element_text(size = 30))
ggsave(paste0('~/Desktop/ccaas/', i, this_ccaa, '_deaths.png'),
width = 1280 / 150,
height = 720 / 150)
}
# Deaths too
for(i in 1:length(ccaas)){
this_ccaa <- ccaas[i]
plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(ccaa == this_ccaa)
roll_curve(plot_data, deaths = T, scales = 'fixed', pop = TRUE) + theme(strip.text = element_text(size = 30))
ggsave(paste0('~/Desktop/ccaas/', i, this_ccaa, '_deaths_pop.png'),
width = 1280 / 150,
height = 720 / 150)
}plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(!ccaa %in% c('Melilla'))
roll_curve(plot_data, scales = 'free_y')
ggsave('~/Desktop/b.png',
width = 1280 / 150,
height = 720 / 150)plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(!ccaa %in% c('Melilla'))
roll_curve(plot_data, deaths = T, scales = 'free_y')
ggsave('~/Desktop/c.png',
width = 1280 / 150,
height = 720 / 150)plot_data <- esp_df %>% mutate(geo = ccaa) %>% filter(!ccaa %in% c('Melilla'))
roll_curve(plot_data, deaths = T, scales = 'fixed')
ggsave('~/Desktop/d.png',
width = 1280 / 150,
height = 720 / 150)plot_data <- df_country %>% filter(country %in% c('Spain', 'Italy', 'Germany', 'France', 'US',
'China', 'Korea, South', 'Japan', 'Singapore')) %>% mutate(geo = country)
roll_curve(plot_data, scales = 'free_y')
ggsave('~/Desktop/z.png',
width = 1280 / 150,
height = 720 / 150)World at large
pd <- df_country %>%
group_by(date) %>%
summarise(n = sum(confirmed_cases)) %>%
filter(date < max(date))
ggplot(data = pd,
aes(x = date,
y = n)) +
geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Cases',
title = 'COVID-19 cases')
ggsave('~/Videos/update/a.png',
width = 1280 / 150,
height = 720 / 150)Error in grid.newpage(): could not open file '/home/joebrew/Videos/update/a.png'China vs world
pd <- df_country %>%
group_by(date,
country = ifelse(country == 'China', 'China', 'Other countries')) %>%
summarise(n = sum(confirmed_cases)) %>%
ungroup %>%
filter(date < max(date))
ggplot(data = pd,
aes(x = date,
y = n,
color = country)) +
geom_line(size = 2) +
# geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Cases',
title = 'COVID-19 cases') +
scale_color_manual(name = '',
values = c('red', 'black')) +
theme(legend.text = element_text(size = 25),
legend.position = 'top')
ggsave('~/Videos/update/b.png',
width = 1280 / 150,
height = 720 / 150)Error in grid.newpage(): could not open file '/home/joebrew/Videos/update/b.png'NOn china only
pd <- df_country %>%
group_by(date,
country = ifelse(country == 'China', 'China', 'Other countries')) %>%
summarise(n = sum(confirmed_cases)) %>%
filter(country == 'Other countries') %>%
ungroup %>%
filter(date < max(date))
ggplot(data = pd,
aes(x = date,
y = n)) +
geom_line(size = 2) +
# geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Cases',
title = 'COVID-19 cases, outside of China') 
ggsave('~/Videos/update/c.png',
width = 1280 / 150,
height = 720 / 150)Error in grid.newpage(): could not open file '/home/joebrew/Videos/update/c.png'Case numbers across countries
plot_day_zero(countries = c('France', 'Germany', 'Italy', 'Spain', 'Switzerland', 'Sweden', 'Norway', 'Netherlands'))
# ggsave('~/Videos/update/d.png',
# width = 1280 / 150,
# height = 720 / 150)World at large - deaths
pd <- df_country %>%
group_by(date) %>%
summarise(n = sum(deaths)) %>%
filter(date < max(date))
ggplot(data = pd,
aes(x = date,
y = n)) +
geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Deaths',
title = 'COVID-19 deaths')
# ggsave('~/Videos/update/e.png',
# width = 1280 / 150,
# height = 720 / 150)China vs world deaths
pd <- df_country %>%
group_by(date,
country = ifelse(country == 'China', 'China', 'Other countries')) %>%
summarise(n = sum(deaths)) %>%
ungroup %>%
filter(date < max(date))
ggplot(data = pd,
aes(x = date,
y = n,
color = country)) +
geom_line(size = 2) +
# geom_point() +
theme_simple() +
labs(x = 'Date',
y = 'Deaths',
title = 'COVID-19 deaths') +
scale_color_manual(name = '',
values = c('red', 'black')) +
theme(legend.text = element_text(size = 25),
legend.position = 'top')
# ggsave('~/Videos/update/f.png',
# width = 1280 / 150,
# height = 720 / 150)Asian hope
plot_day_zero(countries = c('Korea, South', 'Japan', 'China', 'Singapore'))
# ggsave('~/Videos/update/g.png',
# width = 1280 / 150,
# height = 720 / 150)Since trajectories are very unstable when cases are low, we’ll exclude from our analysis the first few days, and will only count as “outbreak” once a country reaches 150 or more cumulative cases.
# Doubling time
n_cases_start = 150
countries = c('Italy', 'Spain', 'France', 'Germany', 'Italy', 'Switzerland', 'Denmark', 'US', 'United Kingdom', 'Norway')
# countries <- sort(unique(df_country$country))
out_list <- curve_list <- list()
counter <- 0
for(i in 1:length(countries)){
message(i)
this_country <- countries[i]
sub_data <-df_country %>% filter(country == this_country)
# Only calculate on countries with n_cases_start or greater cases,
# starting at the first day at n_cases_start or greater
ok <- max(sub_data$confirmed_cases, na.rm = TRUE) >= n_cases_start
if(ok){
counter <- counter + 1
pd <- sub_data %>%
filter(!is.na(confirmed_cases)) %>%
mutate(start_date = min(date[confirmed_cases >= n_cases_start])) %>%
mutate(days_since = date - start_date) %>%
filter(days_since >= 0) %>%
mutate(days_since = as.numeric(days_since))
fit <- lm(log(confirmed_cases) ~ days_since, data = pd)
# plot(pd$days_since, log(pd$cases))
# abline(fit)
## Slope
curve <- fit$coef[2]
# Predict days ahead
fake <- tibble(days_since = seq(0, max(pd$days_since) + 5, by = 1))
fake <- left_join(fake, pd %>% dplyr::select(days_since, confirmed_cases, date))
fake$predicted <- exp(predict(fit, newdata = fake))
# Doubling time
dt <- log(2)/fit$coef[2]
out <- tibble(country = this_country,
doubling_time = dt,
slope = curve)
out_list[[counter]] <- out
curve_list[[counter]] <- fake %>% mutate(country = this_country)
}
}
done <- bind_rows(out_list)
print(done)# A tibble: 10 x 3
country doubling_time slope
<chr> <dbl> <dbl>
1 Italy 3.37 0.206
2 Spain 2.26 0.307
3 France 3.03 0.229
4 Germany 2.71 0.256
5 Italy 3.37 0.206
6 Switzerland 3.02 0.230
7 Denmark 5.90 0.117
8 US 2.35 0.295
9 United Kingdom 3.06 0.227
10 Norway 4.07 0.170curves <- bind_rows(curve_list)
# Get curves back in exponential form
# curves$curve <- exp(curves$curve)
# Join doubling time to curves
joined <- left_join(curves, done)
# Get rid of Italy future (since it's the "leader")
joined <- joined %>%
filter(country != 'Italy' |
date <= (Sys.Date() -1))
# Make long format
long <- joined %>%
dplyr::select(date, days_since, country, confirmed_cases, predicted, doubling_time) %>%
tidyr::gather(key, value, confirmed_cases:predicted) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))The below chart shows the trajectories in terms of number of cases in Europe in red, and the predicted trajectories in black. The black line assumes that the doubling rate will stay constant.
cols <- c('red', 'black')
ggplot(data = long,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = long %>% filter(key != 'Confirmed cases'),
size = 1.2, alpha = 0.8) +
geom_point(data = long %>% filter(key == 'Confirmed cases')) +
geom_line(data = long %>% filter(key == 'Confirmed cases'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >150 cumulative cases',
y = 'Cases',
title = 'COVID-19 CASES: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >150 cumulative cases)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15))
Since Italy is “leading the way”, it’s helpful to also compare each country to Italy. Let’s see that.
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Italy')
ol2 <- joined %>% filter(country == 'Italy') %>% dplyr::rename(Italy = confirmed_cases) %>%
dplyr::select(Italy, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, confirmed_cases, predicted, Italy,doubling_time)
ol <- tidyr::gather(ol, key, value, confirmed_cases: Italy) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
cols <- c('red', 'blue', 'black')
ggplot(data = ol,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = ol %>% filter(!key %in% c('Confirmed cases', 'Italy')),
size = 1.2, alpha = 0.8) +
geom_line(data = ol %>% filter(key %in% c('Italy')),
size = 0.8, alpha = 0.8) +
geom_point(data = ol %>% filter(key == 'Confirmed cases')) +
geom_line(data = ol %>% filter(key == 'Confirmed cases'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >150 cumulative cases',
y = 'Cases',
title = 'COVID-19 CASES: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >150 cumulative cases)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15))
In the above, what’s striking is how many places have trajectories that are worse than Italy’s. Yes, Italy has more cases, but it’s doubling time is less. Either that changes soon, or these other countries will soon have more cases than Italy.
Deaths or cases?
The number of cases is not necessarily the best indicator for the severity of an outbreak of this nature. Why? Because (a) testing rates and protocols are different by place and (b) testing rates are different by time (since health services are changing their approaches as things develop). In other words, when we compare the number of cases by place and time, we are introducing significant bias.
Using deaths to gauge the magnitude of the outbreak is also problematic. Death rates are differential by age, so the number of deaths depends on a country’s population period, or age structure. Also, death rates will be a function of health services, which are not of the same quality every where. And, of course, like cases, we don’t necessarily know about all of the deaths that occur because of COVID-19.
Still, there’s an argument that death rates have less bias than case rates because deaths are easier to identify than cases. Let’s accept that argument, for the time being, and have a look at death rates by country.
# Doubling time
n_deaths_start = 5
countries = c('Italy', 'Spain', 'France', 'Italy', 'Switzerland', 'Denmark', 'US', 'United Kingdom', 'Norway', 'Germany')
# countries <- sort(unique(df_country$country))
make_double_time <- function(data = df_country,
the_country = 'Spain',
n_deaths_start = 5,
time_ahead = 7){
sub_data <-data %>% filter(country == the_country)
# Only calculate on countries with n_cases_start or greater cases,
# starting at the first day at n_cases_start or greater
ok <- max(sub_data$deaths, na.rm = TRUE) >= n_deaths_start
if(ok){
counter <- counter + 1
pd <- sub_data %>%
filter(!is.na(deaths)) %>%
mutate(start_date = min(date[deaths >= n_deaths_start])) %>%
mutate(days_since = date - start_date) %>%
filter(days_since >= 0) %>%
mutate(days_since = as.numeric(days_since)) %>%
mutate(the_weight = 1/(1 + (as.numeric(max(date) - date))))
fit <- lm(log(deaths) ~ days_since,
weights = the_weight,
data = pd)
# fitlo <- loess(deaths ~ days_since, data = pd)
# plot(pd$days_since, log(pd$cases))
# abline(fit)
## Slope
# curve <- fit$coef[2]
# Predict days ahead
day0 <- pd$date[pd$days_since == 0]
fake <- tibble(days_since = seq(0, max(pd$days_since) + time_ahead, by = 1))
fake <- fake %>%mutate(date = seq(day0, day0+max(fake$days_since), by = 1))
fake <- left_join(fake, pd %>% dplyr::select(days_since, deaths, date))
fake$predicted <- exp(predict(fit, newdata = fake))
# fake$predictedlo <- predict(fitlo, newdata = fake)
ci <- exp(predict(fit, newdata = fake, interval = 'prediction'))
# cilo <- predict(fitlo, newdata = fake, interval = 'prediction')
fake$lwr <- ci[,'lwr']
fake$upr <- ci[,'upr']
# fake$lwrlo <- ci[,'lwr']
# fake$uprlo <- ci[,'upr']
# Doubling time
dt <- log(2)/fit$coef[2]
fake %>% mutate(country = the_country) %>% mutate(doubling_time = dt)
}
}
plot_double_time <- function(data, ylog = F){
the_labs <- labs(x = 'Date',
y = 'Deaths',
title = paste0('Predicted deaths in ', data$country[1]))
long <- data %>%
tidyr::gather(key, value, deaths:predicted) %>%
mutate(key = Hmisc::capitalize(key))
g <- ggplot() +
geom_ribbon(data = data %>% filter(date > max(long$date[!is.na(long$value) & long$key == 'Deaths'])),
aes(x = date,
ymax = upr,
ymin = lwr),
alpha =0.6,
fill = 'darkorange') +
geom_line(data = long,
aes(x = date,
y = value,
group = key,
lty = key)) +
geom_point(data = long %>% filter(key == 'Deaths'),
aes(x = date,
y = value)) +
theme_simple() +
theme(legend.position = 'right',
legend.title = element_blank()) +
the_labs
if(ylog){
g <- g + scale_y_log10()
}
return(g)
}
options(scipen = '999')
data <- make_double_time(n_deaths_start = 150, time_ahead = 7)
data# A tibble: 15 x 8
days_since date deaths predicted lwr upr country doubling_time
<dbl> <date> <dbl> <dbl> <dbl> <dbl> <chr> <dbl>
1 0 2020-03-15 288 340. 296. 390. Spain 2.59
2 1 2020-03-16 491 444. 393. 502. Spain 2.59
3 2 2020-03-17 598 581. 521. 647. Spain 2.59
4 3 2020-03-18 767 760. 689. 837. Spain 2.59
5 4 2020-03-19 1002 993. 908. 1086. Spain 2.59
6 5 2020-03-20 1326 1298. 1190. 1416. Spain 2.59
7 6 2020-03-21 1720 1697. 1552. 1855. Spain 2.59
8 7 2020-03-22 2182 2218. 2015. 2442. Spain 2.59
9 8 2020-03-23 NA 2900. 2606. 3227. Spain 2.59
10 9 2020-03-24 NA 3791. 3361. 4276. Spain 2.59
11 10 2020-03-25 NA 4956. 4326. 5677. Spain 2.59
12 11 2020-03-26 NA 6478. 5561. 7547. Spain 2.59
13 12 2020-03-27 NA 8469. 7142. 10042. Spain 2.59
14 13 2020-03-28 NA 11071. 9166. 13372. Spain 2.59
15 14 2020-03-29 NA 14473. 11758. 17815. Spain 2.59dir.create('~/Desktop/ccaa_predictions')
plot_double_time(data, ylog = T) +
labs(subtitle = 'Basic log-linear model weighted at (1 + (1/ days ago)), assuming no change in growth trajectory since first day at >150 deaths')
ggsave('~/Desktop/ccaa_predictions/spain.png')
# All ccaas
ccaas <- sort(unique(esp_df$ccaa))
for(i in 1:length(ccaas)){
message(i)
this_ccaa <- ccaas[i]
sub_data <- esp_df %>% mutate(country = ccaa)
try({
data <- make_double_time(
data = sub_data,
the_country = this_ccaa,
n_deaths_start = 5,
time_ahead = 7)
plot_double_time(data, ylog = T) +
labs(subtitle = 'Basic log-linear model weighted at (1 + (1/ days ago)), assuming no change in growth trajectory since first day at >5 deaths')
ggsave(paste0('~/Desktop/ccaa_predictions/',
this_ccaa, '.png'),
height = 4.9,
width = 8.5)
})
}Error in UseMethod("gather_") :
no applicable method for 'gather_' applied to an object of class "NULL"Error in UseMethod("gather_") :
no applicable method for 'gather_' applied to an object of class "NULL"
Error in UseMethod("gather_") :
no applicable method for 'gather_' applied to an object of class "NULL"# all_countries <- sort(unique(df_country$country))
# for(i in 1:length(all_countries)){
# this_country <- all_countries[i]
# data <- make_double_time(the_country = this_country, n_deaths_start = 5)
# if(!is.null(data)){
# # print(this_country)
# g <- plot_double_time(data, ylog = F) +
# labs(subtitle = 'Basic log-linear model assuming no change in growth trajectory since first day at >5 deaths')
# ggsave(paste0('~/Desktop/', this_country, '.png'), height = 5, width = 8)
# print(data)
# }
# }out_list <- curve_list <- list()
counter <- 0
for(i in 1:length(countries)){
message(i)
this_country <- countries[i]
sub_data <-df_country %>% filter(country == this_country)
# Only calculate on countries with n_cases_start or greater cases,
# starting at the first day at n_cases_start or greater
ok <- max(sub_data$deaths, na.rm = TRUE) >= n_deaths_start
if(ok){
counter <- counter + 1
pd <- sub_data %>%
filter(!is.na(deaths)) %>%
mutate(start_date = min(date[deaths >= n_deaths_start])) %>%
mutate(days_since = date - start_date) %>%
filter(days_since >= 0) %>%
mutate(days_since = as.numeric(days_since))
fit <- lm(log(deaths) ~ days_since, data = pd)
# plot(pd$days_since, log(pd$cases))
# abline(fit)
## Slope
# curve <- fit$coef[2]
# Predict days ahead
fake <- tibble(days_since = seq(0, max(pd$days_since) + 5, by = 1))
fake <- left_join(fake, pd %>% dplyr::select(days_since, deaths, date))
fake$predicted <- exp(predict(fit, newdata = fake))
# Doubling time
dt <- log(2)/fit$coef[2]
out <- tibble(country = this_country,
doubling_time = dt)
out_list[[counter]] <- out
curve_list[[counter]] <- fake %>% mutate(country = this_country)
}
}Error in lm.fit(x, y, offset = offset, singular.ok = singular.ok, ...): NA/NaN/Inf in 'y'done <- bind_rows(out_list)
curves <- bind_rows(curve_list)
# Get curves back in exponential form
# curves$curve <- exp(curves$curve)
# Join doubling time to curves
joined <- left_join(curves, done)
# Get rid of Italy future (since it's the "leader")
joined <- joined %>%
filter(country != 'Italy' |
date <= (Sys.Date() -1))
# Make long format
long <- joined %>%
dplyr::select(date, days_since, country, deaths, predicted, doubling_time) %>%
tidyr::gather(key, value, deaths:predicted) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))cols <- c('red', 'black')
sub_data <- long %>% filter(country != 'US')
ggplot(data = sub_data,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = sub_data %>% filter(key != 'Deaths'),
size = 1.2, alpha = 0.8) +
geom_point(data = sub_data %>% filter(key == 'Deaths')) +
geom_line(data = sub_data %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 cumulative deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15))
Let’s again overlay Italy.
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Italy')
ol2 <- joined %>% filter(country == 'Italy') %>% dplyr::rename(Italy = deaths) %>%
dplyr::select(Italy, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, deaths, predicted, Italy,doubling_time)
ol <- tidyr::gather(ol, key, value, deaths: Italy) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
cols <- c('red', 'blue', 'black')
sub_data <- ol %>%
filter(!(key == 'Predicted (based on current doubling time)' &
country == 'Spain' &
days_since > 13))
ggplot(data = sub_data,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = sub_data %>% filter(!key %in% c('Deaths', 'Italy')),
size = 1.2, alpha = 0.8) +
geom_line(data = sub_data %>% filter(key %in% c('Italy')),
size = 0.8, alpha = 0.8) +
geom_point(data = sub_data %>% filter(key == 'Deaths')) +
geom_line(data = sub_data %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,2)) +
scale_color_manual(name = '',
values = cols) +
scale_y_log10() +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15)) 
Let’s look just at Spain
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Italy',
country == 'Spain')
ol2 <- joined %>% filter(country == 'Italy') %>% dplyr::rename(Italy = deaths) %>%
dplyr::select(Italy, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, deaths, predicted, Italy,doubling_time)
ol <- tidyr::gather(ol, key, value, deaths: Italy) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)',
ifelse(key == 'Deaths', 'Spain', key)))
cols <- c('blue', 'black', 'red')
ggplot(data = ol,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = ol %>% filter(!key %in% c('Deaths', 'Italy')),
size = 1.2, alpha = 0.8) +
geom_line(data = ol %>% filter(key %in% c('Italy')),
size = 0.8, alpha = 0.8) +
# geom_point(data = ol %>% filter(key == 'Deaths')) +
geom_point(data = ol %>% filter(country == 'Spain',
key == 'Spain'), size = 4, alpha = 0.6) +
geom_line(data = ol %>% filter(key == 'Deaths'),
size = 0.8) +
# facet_wrap(~paste0(country, '\n',
# '(doubling time: ',
# round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,1)) +
scale_color_manual(name = '',
values = cols) +
scale_y_log10() +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = 13),
plot.title = element_text(size = 15),
axis.title = element_text(size = 18))
The importance of lag
Things are changing very rapidly. And measures being taken by these countries will have an impact on the outbreak.
But it’s important to remember that there is a lag between when an intervention takes place and when its effect is notable. Because of the incubation period - the number of days between someone getting infected and becoming sick - what we do today won’t really have an effect until next weekend. And the clinical cases that present today are among people who got infected a week ago.
Disease control measures work. We can see that clearly in the case of Hubei, Wuhan, Iran, Japan. And they will work in Europe too. But because many of these measures were implemented very recently, we won’t likely see a major effect for at least a few more days.
In the mean time, it’s important to practice social distancing. Stay away from others to keep both you and others safe. Listen to Health Authorities. Take this very seriously.
Spain and Italy regions
# Madrid vs Lombardy deaths
n_death_start <- 5
pd <- esp_df %>%
# filter(ccaa == 'Madrid') %>%
dplyr::select(date, ccaa, cases, deaths) %>%
bind_rows(ita %>%
# filter(ccaa == 'Lombardia') %>%
dplyr::select(date, ccaa, cases, deaths)) %>%
arrange(date) %>%
group_by(ccaa) %>%
mutate(first_n_death = min(date[deaths >= n_death_start])) %>%
ungroup %>%
mutate(days_since_n_deaths = date - first_n_death) %>%
filter(is.finite(days_since_n_deaths))
pd$country <- pd$ccaa
pd$confirmed_cases <- pd$cases
countries <- sort(unique(pd$country))
out_list <- curve_list <- list()
counter <- 0
for(i in 1:length(countries)){
message(i)
this_country <- countries[i]
sub_data <- pd %>% filter(country == this_country)
# Only calculate on countries with n_cases_start or greater cases,
# starting at the first day at n_cases_start or greater
# ok <- max(sub_data$deaths, na.rm = TRUE) >= n_deaths_start
ok <- length(which(sub_data$deaths >= n_deaths_start))
if(ok){
counter <- counter + 1
sub_pd <- sub_data %>%
filter(!is.na(deaths)) %>%
mutate(start_date = min(date[deaths >= n_deaths_start])) %>%
mutate(days_since = date - start_date) %>%
filter(days_since >= 0) %>%
mutate(days_since = as.numeric(days_since))
fit <- lm(log(deaths) ~ days_since, data = sub_pd)
# plot(pd$days_since, log(pd$cases))
# abline(fit)
## Slope
# curve <- fit$coef[2]
# Predict days ahead
fake <- tibble(days_since = seq(0, max(sub_pd$days_since) + 5, by = 1))
fake <- left_join(fake, sub_pd %>% dplyr::select(days_since, deaths, date))
fake$predicted <- exp(predict(fit, newdata = fake))
# Doubling time
dt <- log(2)/fit$coef[2]
out <- tibble(country = this_country,
doubling_time = dt)
out_list[[counter]] <- out
curve_list[[counter]] <- fake %>% mutate(country = this_country)
}
}
done <- bind_rows(out_list)
curves <- bind_rows(curve_list)
# Get curves back in exponential form
# curves$curve <- exp(curves$curve)
# Join doubling time to curves
joined <- left_join(curves, done)
# Make long format
long <- joined %>%
dplyr::select(date, days_since, country, deaths, predicted, doubling_time) %>%
tidyr::gather(key, value, deaths:predicted) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
# Remove those with not enough data to have a doubling time yet
long <- long %>% filter(!is.na(doubling_time))text_size <- 12
cols <- c('red', 'black')
ggplot(data = long,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
geom_line(data = long %>% filter(key != 'Deaths'),
size = 1.2, alpha = 0.8) +
geom_point(data = long %>% filter(key == 'Deaths')) +
geom_line(data = long %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_y_log10() +
scale_linetype_manual(name ='',
values = c(1,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >150 cumulative cases',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = text_size * 0.5),
plot.title = element_text(size = 15))
Let’s overlay Lombardy
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Lombardia')
ol2 <- joined %>% filter(country == 'Lombardia') %>% dplyr::rename(Lombardia = deaths) %>%
dplyr::select(Lombardia, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, deaths, predicted, Lombardia,doubling_time)
ol <- tidyr::gather(ol, key, value, deaths: Lombardia) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
# Remove those with not enough data to have a doubling time yet
ol <- ol %>% filter(!is.na(doubling_time))
cols <- c('red', 'blue', 'black')
ggplot(data = ol,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
scale_y_log10() +
geom_line(data = ol %>% filter(!key %in% c('Deaths', 'Italy')),
size = 1.2, alpha = 0.8) +
geom_line(data = ol %>% filter(key %in% c('Lombardia')),
size = 0.5, alpha = 0.8) +
geom_point(data = ol %>% filter(key == 'Deaths')) +
geom_line(data = ol %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = text_size * 0.5),
plot.title = element_text(size = 15))
Show only Spanish regions vs. Lombardy
text_size <- 14
# Overlay Italy
ol1 <- joined %>% filter(!country %in% 'Lombardia')
ol2 <- joined %>% filter(country == 'Lombardia') %>% dplyr::rename(Lombardia = deaths) %>%
dplyr::select(Lombardia, days_since)
ol <- left_join(ol1, ol2) %>%
dplyr::select(days_since, date, country, deaths, predicted, Lombardia,doubling_time)
ol <- tidyr::gather(ol, key, value, deaths: Lombardia) %>%
mutate(key = Hmisc::capitalize(gsub('_', ' ', key))) %>%
mutate(key = ifelse(key == 'Predicted', 'Predicted (based on current doubling time)', key))
# Remove those with not enough data to have a doubling time yet
ol <- ol %>% filter(!is.na(doubling_time))
# Only Spain
ol <- ol %>% filter(country %in% esp_df$ccaa) %>%
filter(!country %in% 'Aragón')
cols <- c('red', 'blue', 'black')
ggplot(data = ol,
aes(x = days_since,
y = value,
lty = key,
color = key)) +
scale_y_log10() +
geom_line(data = ol %>% filter(!key %in% c('Deaths', 'Lombardia')),
size = 1.2, alpha = 0.8) +
geom_line(data = ol %>% filter(key %in% c('Lombardia')),
size = 0.5, alpha = 0.8) +
geom_point(data = ol %>% filter(key == 'Deaths')) +
geom_line(data = ol %>% filter(key == 'Deaths'),
size = 0.8) +
facet_wrap(~paste0(country, '\n',
'(doubling time: ',
round(doubling_time, digits = 1), ' days)'), scales = 'free') +
theme_simple() +
scale_linetype_manual(name ='',
values = c(1,6,2)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
labs(x = 'Days since first day at >5 deaths',
y = 'Deaths',
title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
theme(strip.text = element_text(size = text_size * 0.6),
plot.title = element_text(size = 15))
Same plot but overlayed
Same as above, but overlaid
text_size <-10
# cols <- c('red', 'black')
long <- long %>% filter(country %in% c('Lombardia',
'Emilia Romagna') |
country %in% esp_df$ccaa) %>%
filter(country != 'Aragón')
places <- sort(unique(long$country))
cols <- colorRampPalette(RColorBrewer::brewer.pal(n = 7, 'Spectral'))(length(places))
cols[which(places == 'Madrid')] <- 'red'
cols[which(places == 'Cataluña')] <- 'purple'
cols[which(places == 'Lombardia')] <- 'darkorange'
cols[which(places == 'Emilia Romagna')] <- 'darkgreen'
long$key <- ifelse(long$key != 'Deaths', 'Predicted', long$key)
long$key <- ifelse(long$key == 'Predicted', 'Muertes\nprevistas',
'Muertes\nobservadas')
# Keep only Madrid, Lombardy, Emilia Romagna
long <- long %>%
filter(country %in% c('Madrid',
'Lombardia',
'Emilia Romagna'))
ggplot(data = long,
aes(x = days_since,
y = value,
lty = key,
color = country)) +
geom_point(data = long %>% filter(key == 'Muertes\nobservadas'), size = 2, alpha = 0.8) +
geom_line(data = long %>% filter(key == 'Muertes\nprevistas'), size = 1, alpha = 0.7) +
geom_line(data = long %>% filter(key != 'Muertes\nprevistas'), size = 0.8) +
theme_simple() +
scale_y_log10() +
scale_linetype_manual(name ='',
values = c(1,4)) +
scale_color_manual(name = '',
values = cols) +
theme(legend.position = 'top') +
# labs(x = 'Days since first day at 5 or more cumulative deaths',
# y = 'Deaths',
# title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
# caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
# subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
labs(x = 'Días desde el primer día a 5 o más muertes acumuladas',
y = 'Muertes (escala logarítmica)',
title = 'Muertes por COVID-19',
caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
subtitle = '(Tasa de crecimiento calculada desde el primer día a 5 o más muertes acumuladas)\n(Muertes "previstas": suponiendo que no hay cambios en la tasa de crecimiento)') +
theme(strip.text = element_text(size = text_size * 0.75),
plot.title = element_text(size = text_size * 3),
legend.text = element_text(size = text_size * 1.5),
axis.title = element_text(size = text_size * 2),
axis.text = element_text(size = text_size * 2))
# cols <- c(cols, 'darkorange')
# ggplot(data = ol,
# aes(x = days_since,
# y = value,
# lty = key,
# color = key)) +
# scale_y_log10() +
# geom_line(aes(color = country)) +
#
# # geom_line(data = ol %>% filter(!key %in% c('Deaths', 'Italy')),
# # size = 1.2, alpha = 0.8) +
# # geom_line(data = ol %>% filter(key %in% c('Lombardia')),
# # size = 0.5, alpha = 0.8) +
# # geom_point(data = ol %>% filter(key == 'Deaths')) +
# # geom_line(data = ol %>% filter(key == 'Deaths'),
# # size = 0.8) +
# theme_simple() +
# scale_linetype_manual(name ='',
# values = c(1,6,2)) +
# scale_color_manual(name = '',
# values = cols) +
# theme(legend.position = 'top') +
# labs(x = 'Days since first day at >5 deaths',
# y = 'Deaths',
# title = 'COVID-19 DEATHS: ("predicted" assumes no change in doubling time)',
# caption = 'Data from Johns Hopkins. Processing: Joe Brew @joethebrew. Code: github.com/databrew/covid19',
# subtitle = '(Doubling time calculated since first day at >5 cumulative deaths)') +
# theme(strip.text = element_text(size = text_size * 1),
# plot.title = element_text(size = 15))# Map data preparation
if(!'map.RData' %in% dir()){
esp1 <- getData(name = 'GADM', country = 'ESP', level = 1)
# Remove canary
esp1 <- esp1[esp1@data$NAME_1 != 'Islas Canarias',]
espf <- fortify(esp1, region = 'NAME_1')
# Standardize names
# Convert names
map_names <- esp1@data$NAME_1
data_names <- sort(unique(esp_df$ccaa))
names_df <- tibble(NAME_1 = c('Andalucía',
'Aragón',
'Cantabria',
'Castilla-La Mancha',
'Castilla y León',
'Cataluña',
'Ceuta y Melilla',
'Comunidad de Madrid',
'Comunidad Foral de Navarra',
'Comunidad Valenciana',
'Extremadura',
'Galicia',
'Islas Baleares',
'La Rioja',
'País Vasco',
'Principado de Asturias',
'Región de Murcia'),
ccaa = c('Andalucía',
'Aragón',
'Cantabria',
'CLM',
'CyL',
'Cataluña',
'Melilla',
'Madrid',
'Navarra',
'C. Valenciana',
'Extremadura',
'Galicia',
'Baleares',
'La Rioja',
'País Vasco',
'Asturias',
'Murcia'))
espf <- left_join(espf %>% dplyr::rename(NAME_1 = id), names_df)
centroids <- data.frame(coordinates(esp1))
names(centroids) <- c('long', 'lat')
centroids$NAME_1 <- esp1$NAME_1
centroids <- centroids %>% left_join(names_df)
# Get random sampling points
random_list <- list()
for(i in 1:nrow(esp1)){
message(i)
shp <- esp1[i,]
# bb <- bbox(shp)
this_ccaa <- esp1@data$NAME_1[i]
# xs <- runif(n = 500, min = bb[1,1], max = bb[1,2])
# ys <- runif(n = 500, min = bb[2,1], max = bb[2,2])
# random_points <- expand.grid(long = xs, lat = ys) %>%
# mutate(x = long,
# y = lat)
# coordinates(random_points) <- ~x+y
# proj4string(random_points) <- proj4string(shp)
# get ccaa
message('getting locations of randomly generated points')
# polys <- over(random_points,polygons(shp))
# polys <- as.numeric(polys)
random_points <- spsample(shp, n = 20000, type = 'random')
random_points <- data.frame(random_points)
random_points$NAME_1 <- this_ccaa
random_points <- left_join(random_points, names_df) %>% dplyr::select(-NAME_1)
random_list[[i]] <- random_points
}
random_points <- bind_rows(random_list)
random_points <- random_points %>% mutate(long = x,
lat = y)
save(espf,
esp1,
names_df,
centroids,
random_points,
file = 'map.RData')
} else {
load('map.RData')
}
# Define a function for adding zerio
add_zero <-
function (x, n)
{
x <- as.character(x)
adders <- n - nchar(x)
adders <- ifelse(adders < 0, 0, adders)
for (i in 1:length(x)) {
if (!is.na(x[i])) {
x[i] <- paste0(paste0(rep("0", adders[i]), collapse = ""),
x[i], collapse = "")
}
}
return(x)
}remake_world_map <- FALSE
options(scipen = '999')
if(remake_world_map){
# World map animation
world <- map_data('world')
# world <- ne_countries(scale = "medium", returnclass = "sf")
# Get plotting data
pd <- df_country %>%
dplyr::select(date, lng, lat, n = confirmed_cases)
dates <- sort(unique(pd$date))
n_days <- length(dates)
# # Define vectors for projection
# vec_lon <- seq(30, -20, length = n_days)
# vec_lat <- seq(25, 15, length = n_days)
dir.create('animation')
for(i in 1:n_days){
message(i, ' of ', n_days)
this_date <- dates[i]
# this_lon <- vec_lon[i]
# this_lat <- vec_lat[i]
# the_crs <-
# paste0("+proj=laea +lat_0=", this_lat,
# " +lon_0=",
# this_lon,
# " +x_0=4321000 +y_0=3210000 +ellps=GRS80 +units=m +no_defs ")
sub_data <- pd %>%
filter(date == this_date)
# coordinates(sub_data) <- ~lng+lat
# proj4string(sub_data) <- proj4string(esp1)
# # sub_data <- spTransform(sub_data,
# # the_crs)
# coordy <- coordinates(sub_data)
# sub_data@data$long <- coordy[,1]
# sub_data@data$lat <- coordy[,2]
g <- ggplot() +
geom_polygon(data = world,
aes(x = long,
y = lat,
group = group),
fill = 'black',
color = 'white',
size = 0.1) +
theme_map() +
geom_point(data = sub_data %>% filter(n > 0) %>% mutate(Deaths = n),
aes(x = lng,
y = lat,
size = Deaths),
color = 'red',
alpha = 0.6) +
geom_point(data = tibble(x = c(0,0), y = c(0,0), Deaths = c(1, 100000)),
aes(x = x,
y = y,
size = Deaths),
color = 'red',
alpha =0.001) +
scale_size_area(name = '', breaks = c(100, 1000, 10000, 100000),
max_size = 25
) +
# scale_size_area(name = '', limits = c(1, 10), breaks = c(0, 10, 30, 50, 70, 100, 200, 500)) +
labs(title = this_date) +
theme(plot.title = element_text(size = 30),
legend.text = element_text(size = 15),
legend.position = 'left')
plot_number <- add_zero(i, 3)
ggsave(filename = paste0('animation/', plot_number, '.png'),
plot = g,
width = 9.5,
height = 5.1)
}
setwd('animation')
system('convert -delay 30x100 -loop 0 *.png result.gif')
setwd('..')
}Maps of Spain
make_map <- function(var = 'deaths',
date = NULL,
pop = FALSE,
pop_factor = 100000,
points = FALSE,
line_color = 'white'){
if(is.null(date)){
the_date <- max(esp_df$date)
} else {
the_date <- date
}
left <- espf
right <- esp_df[esp_df$date == the_date,c('ccaa', var)]
names(right)[ncol(right)] <- 'var'
if(pop){
right <- left_join(right, esp_pop)
right$var <- right$var / right$pop * pop_factor
}
map <- left_join(left, right)
if(points){
the_points <- centroids %>%
left_join(right)
g <- ggplot() +
geom_polygon(data = map,
aes(x = long,
y = lat,
group = group),
fill = 'black',
color = line_color,
lwd = 0.4, alpha = 0.8) +
geom_point(data = the_points,
aes(x = long,
y = lat,
size = var),
color = 'red',
alpha = 0.7) +
scale_size_area(name = '', max_size = 20)
} else {
# cols <- c('#008080','#70a494','#b4c8a8','#f6edbd','#edbb8a','#de8a5a','#ca562c')
cols <- RColorBrewer::brewer.pal(n = 8, name = 'Blues')
g <- ggplot(data = map,
aes(x = long,
y = lat,
group = group)) +
geom_polygon(aes(fill = var),
lwd = 0.3,
color = line_color) +
scale_fill_gradientn(name = '',
colours = cols)
# scale_fill_viridis(name = '' ,option = 'magma',
# direction = -1)
}
g +
theme_map() +
labs(subtitle = paste0('Data as of ', max(esp_df$date)))
}
make_dot_map <- function(var = 'deaths',
date = NULL,
pop = FALSE,
pop_factor = 100,
point_factor = 1,
points = FALSE,
point_color = 'darkred',
point_size = 0.6,
point_alpha = 0.5){
if(is.null(date)){
the_date <- max(esp_df$date)
} else {
the_date <- date
}
right <- esp_df[esp_df$date == the_date,c('ccaa', var)]
names(right)[ncol(right)] <- 'var'
if(pop){
right <- left_join(right, esp_pop)
right$var <- right$var / right$pop * pop_factor
}
map_data <- esp1@data %>%
left_join(names_df) %>%
left_join(right)
map_data$var <- map_data$var / point_factor
out_list <- list()
for(i in 1:nrow(map_data)){
sub_data <- map_data[i,]
this_value = round(sub_data$var)
if(this_value >= 1){
this_ccaa = sub_data$ccaa
# get some points
sub_points <- random_points %>% filter(ccaa == this_ccaa)
sampled_points <- sub_points %>% dplyr::sample_n(this_value)
out_list[[i]] <- sampled_points
}
}
the_points <- bind_rows(out_list)
g <- ggplot() +
geom_polygon(data = espf,
aes(x = long,
y = lat,
group = group),
fill = 'white',
color = 'black',
lwd = 0.4, alpha = 0.8) +
geom_point(data = the_points,
aes(x = long,
y = lat),
color = point_color,
size = point_size,
alpha = point_alpha)
g +
theme_map() +
labs(subtitle = paste0('Data as of ', max(esp_df$date)))
}Deaths
Absolute number of deaths: points
make_map(var = 'deaths',
points = T) +
labs(title = 'Number of deaths',
caption = '@joethebrew')
Absolute number of deaths: choropleth
make_map(var = 'deaths',
line_color = 'darkgrey',
points = F) +
labs(title = 'Number of deaths',
caption = '@joethebrew')
Number of deaths adjusted by population: points
make_map(var = 'deaths', pop = TRUE, points = T) +
labs(title = 'Number of deaths per 100,000',
caption = '@joethebrew')
Number of deaths adjusted by population: polygons
make_map(var = 'deaths', pop = TRUE, points = F, line_color = 'darkgrey') +
labs(title = 'Number of deaths per 100,000',
caption = '@joethebrew')
Number of deaths: 1 dot per death
make_dot_map(var = 'deaths', point_size = 0.05) +
labs(title = 'COVID-19 deaths: 1 point = 1 death\nImportant: points are random within each CCAA; do not reflect exact location',
caption = '@joethebrew')
Cases
Absolute number of cases: points
make_map(var = 'confirmed_cases',
points = T) +
labs(title = 'Number of confirmed cases',
caption = '@joethebrew')
Absolute number of cases: choropleth
make_map(var = 'confirmed_cases',
line_color = 'darkgrey',
points = F) +
labs(title = 'Number of confirmed cases',
caption = '@joethebrew')
Number of cases adjusted by population: points
make_map(var = 'confirmed_cases', pop = TRUE, points = T) +
labs(title = 'Number of confirmed cases per 100,000',
caption = '@joethebrew')
Number of cases adjusted by population: polygons
make_map(var = 'confirmed_cases', pop = TRUE, points = F,
line_color = 'darkgrey') +
labs(title = 'Number of confirmed cases per 100,000',
caption = '@joethebrew')
Number of cases: points
make_dot_map(var = 'confirmed_cases',
point_size = 0.05, point_alpha = 0.5, point_factor = 10) +
labs(title = 'COVID-19 cases: 1 point = 10 cases\nImportant: points are random within each CCAA; do not reflect exact location',
caption = '@joethebrew')